: Stereopsis, an important source of information about physical distance, depends on the human ability to respond to binocular disparity. How is disparity encoded by the human visual system? A substantial body of electrophysiological and psychophysical data shows that there is a special disparity system in the primate cortex, but the exact mechanism for combining the two monocular signals is as yet unknown. We propose a series of psychophysical experiments to test a mechanistic model of a "primitive" disparity system, a model for simple local detectors found in early stages of visual processing. Based on the vast knowledge of contrast sensitivity and spatial vision accumulated over the past three decades, this model can account for previous psychophysical measurements of stereoacuity and disparity discrimination, including some generated by this laboratory, and it does mimic the tuning characteristics of single cortical units that are sensitive to disparity. This model does not solve the "binocular correspondence" problem, nor does it account for higher order interactions occurring between local disparity detectors across the visual field. Additional experiments will examine the effects of these higher order interactions. Disparity is often thought to be useful in breaking camouflage as in Julesz' famous random dot stereograms. How well does disparity per se define visual forms? Gradual changes in disparity (smooth surfaces tilted in depth) can be very difficult to see. Is this because of inhibitory interactions between adjacent detectors? A final series of experiments will explore the relationship between postulated binocular and monocular mechanisms. We use measurements of spatial localization (hyperacuity) to study the competition between binocular and monocular signals in normal observers. In stereoanomalous and amblyopic observers, there are significant deficiencies in binocular processing probably due to an imbalance between the monocular signals coming from the two eyes. Therefore, experiments are planned to study the binocular-monocular competition in stereoanomalous observers.